#1 and #9 seem really important. (Managers must be given actual control over a project, contractors must be allowed to test the work they do.)
I've seen how destructive not having those two policies in play can be, even in a small organization. You have to give the people on charge the authority and ability to get things done, and you have to make sure the people who do the work get the chance to see firsthand whether or not the work "flies" in the real world.
It is amazing how relevant most of these rules are more than 50 years after they were published. Kelly would have been one heck of a founder.
I found this one on documentation particularly applicable.. 5. There must be a minimum number of reports required, but important work must be recorded thoroughly.
Lockheed Martin's Skunk Works were responsible for some of the most innovative aircraft designs from WWII. These were the 14 rules put in place at the birth of the Skunk Works.
Looking past the military stuff, some of them are quite applicable to startups even to this day.
There was an unofficial 15th rule: Don't work with the Navy. (From the book Skunk Works by Ben Rich who succeeded Kelly. Great book. Read it.)
"14. Because only a few people will be used in engineering and most other areas, ways must be provided to reward good performance by pay not based on the number of personnel supervised."
This is something large corporations struggle with today. The prevalence of of startups seems to indicate that there isn't a good alternative to the track of promoting good engineers to management. Engineers who want to do design work don't want to be managers.
I'm reminded of a charter school I read about in the NYT(don't have the link sorry) that was unusual for paying its best teachers more than the principal. They had a principal who understood that even though he was the teachers boss he needed to attract good teachers with high salary.